Search Service Metrics
- Capella Operational
A list of the metrics provided by the Search Service.
|
Tip
|
Average latency per query, using gRPC for streaming, for an index |
Average latency of inter-node queries per unit time for an index |
Average latency of queries per unit time for an index |
Total number of bytes from batches yet to be indexed. |
Total number of bytes from batches which have been indexed. |
The time the service booted in fractional seconds since Unix epoch. |
The number of distinct operations recording Compute Units (CUs) with Regulator. |
The number of distinct operations recording Read Units (RUs) with Regulator. |
The number of distinct operations recording Write Units (WUs) with Regulator. |
The number of Compute Units (CUs) refunded. |
The number of Read Units (RUs) refunded. |
The number of Write Units (WUs) refunded. |
The difference between the number of batches that have been indexed and the number of batches yet to be indexed |
Number of documents in the index |
The number of query timer events received at the global query endpoint |
Mean runtime for queries received at the global query endpoint |
Median runtime for queries received at the global query endpoint |
80th percentile runtime for queries received at the global query endpoint |
99th percentile runtime for queries received at the global query endpoint |
The number of query timer events received at the gRPC endpoint |
Mean runtime for queries received at the gRPC endpoint |
Median runtime for queries received at the gRPC endpoint |
80th percentile runtime for queries received at the gRPC endpoint |
99th percentile runtime for queries received at the gRPC endpoint |
The number of Compute Units (CUs) recorded. |
The number of Read Units (RUs) recorded. |
The number of Write Units (WUs) recorded. |
Total number of batches introduced as part of indexing. |
The number of bytes allocated by the cluster manager as maximum usable memory for fts service |
The number of bytes used on disk by the index |
The number of bytes used on disk by the root segment |
The number of bytes used in memory |
The number of files on disk for an index |
Number of search indexes in the cluster |
Total number of search requests with KNN |
DCP sequence numbers yet to be indexed for an index |
Total number of pindexes currently |
Planned/expected number of pindexes |
The number of entries (terms, records, dictionary rows, etc) by Bleve not yet persisted to storage |
The number of file segments in the root segment |
The number of memory segments in the root segment |
Number of search indexes in the cluster that have vector/vector_base64 fields |
The number of distinct operations recorded with Regulator. |
The percentage of CPU time spent by an index in garbage collection |
The percentage of RAM quota used by the fts service |
The number of times Regulator instructed an operation to be rejected. |
The number of query timer events received at the scoped query endpoint |
Mean runtime for queries received at the scoped query endpoint |
Median runtime for queries received at the scoped query endpoint |
80th percentile runtime for queries received at the scoped query endpoint |
99th percentile runtime for queries received at the scoped query endpoint |
The number of times Regulator instructed an operation to throttle. |
The total time spent throttling (in seconds). |
Total number of batches executed due to the batch size being greater than the maximum number of operations per batch |
Total number of batches executed at regular intervals |
Total number of times Bleve destinations closed |
The number of times Bleve destinations opened |
Total number of gRPC listeners closed |
Total number of gRPC listeners opened |
Total number of gRPC queries rejected due to the memory quota being lesser than the estimated memory required for merging search results from all partitions from the query |
Total number of gRPC SSL listeners closed |
Total number of gRPC SSL listeners opened |
Total number of HTTP limit listeners closed |
Total number of HTTP limit listeners opened |
Total number of HTTPS limit listeners closed |
Total number of HTTPS limit listeners opened |
Total number of queries rejected due to the memory quota being lesser than the estimated memory required for merging search results from all partitions from the query |
Total number of remote(i.e. different node) gRPC requests |
Total number of remote(i.e. different node) gRPC SSL requests when adding clients. |
Total number of remote(i.e. different node) gRPC SSL requests when adding clients. |
Total number of remote(i.e. different node) HTTP requests |
Total number of remote(i.e. different node) HTTP SSL requests |
Total number of remote(i.e. different node) HTTP SSL requests |
Rate of bytes indexed for an index |
Size of results coming back from full text queries for search results, including the entire size of the JSON sent |
Number of bytes written to disk as a result of compaction |
The number of garbage collection events triggered |
The number of internal gRPC requests from the co-ordinating node for the query to other nodes, for an index |
The total number of queries, using gRPC for streaming, for an index |
The number of queries that resulted in an error, using gRPC for streaming, for an index |
The number of queries in the slow query log, using gRPC for streaming, for an index |
The number of queries that exceeded the timeout, using gRPC for streaming, for an index |
The number of internal queries from the co-ordinating node to other nodes, per unit time for an index |
Total bleve knn search operations |
Total number of mutations that qualify for any document filter in an index (in-memory only stat) |
The number of full text queries per second for an index |
The number of FTS queries that resulted in an error due to a bad request |
The number of FTS queries that resulted in an error due to a failure to meet consistency requirements |
The number of FTS queries for an index that resulted in an error |
The number of FTS queries that resulted in an error due to exceeding the maximum result window |
The number of FTS queries that resulted in an error due to only partial results being returned |
The number of queries rejected by the app herder when the memory used exceeds the application's query quota |
The number of FTS queries that resulted in an error while searching in context |
The number of FTS queries in the slow query log |
The number of FTS queries for an index that exceeded the timeout |
Total number of searches served by the active partitions of an index on a node |
Total number of searches served by the replica partitions of an index on a node |
Total time spent processing query requests for an index |
Total bleve synonym search operations |
Number of bleve term searchers |
Total term searchers that have finished serving a query |
The total number of vectors indexed |
Total number of http query requests with KNN rejected by the throttler |